Semi-Supervised Phone Classification using Deep Neural Networks and Stochastic Graph-Based Entropic Regularization

TL;DR

This paper introduces a scalable semi-supervised deep learning method using stochastic graph-based entropic regularization, significantly improving phone classification accuracy with limited labeled data.

Contribution

It presents a novel, efficient stochastic graph regularization technique compatible with mini-batch training for deep neural networks.

Findings

Improves classification accuracy with low labeled data

Competitive performance in fully labeled scenarios

Scalable to large datasets

Abstract

We describe a graph-based semi-supervised learning framework in the context of deep neural networks that uses a graph-based entropic regularizer to favor smooth solutions over a graph induced by the data. The main contribution of this work is a computationally efficient, stochastic graph-regularization technique that uses mini-batches that are consistent with the graph structure, but also provides enough stochasticity (in terms of mini-batch data diversity) for convergence of stochastic gradient descent methods to good solutions. For this work, we focus on results of frame-level phone classification accuracy on the TIMIT speech corpus but our method is general and scalable to much larger data sets. Results indicate that our method significantly improves classification accuracy compared to the fully-supervised case when the fraction of labeled data is low, and it is competitive with other methods in the fully labeled case.